feat: generate Jamo conversion lookup table

WieeRd · Feb 12, 2024 · c8a36dc · c8a36dc
1 parent 20919bb
commit c8a36dc
Show file tree

Hide file tree

Showing 2 changed files with 18 additions and 294 deletions.
diff --git a/mklookup.py b/mklookup.py
@@ -4,8 +4,6 @@
 from collections.abc import Callable
 from typing import TypeVar
 
-# FIX: moum to jungseong can be 1:1 mapped; it does not need a lookup table
-
 
 def jamo_to_compat_jamo(jamo: str, /) -> str | None:
     """Maps a Jamo character to a Compatibility Jamo character."""
@@ -25,12 +23,6 @@ def compat_jaum_to_choseong(compat_jaum: str, /) -> str | None:
         return None
 
 
-def compat_moum_to_jungseong(compat_moum: str, /) -> str:
-    """Maps a Compatibility Moum character to a Jamo Jungseong character."""
-    name = ud.name(compat_moum).split(" ")[-1]  # HANGUL LETTER "A"
-    return ud.lookup(f"HANGUL JUNGSEONG {name}")
-
-
 def compat_jaum_to_jongseong(compat_jaum: str, /) -> str:
     """Maps a Compatibility Jaum character to a Jamo Jongseong character."""
     name = ud.name(compat_jaum).split(" ")[-1]  # HANGUL LETTER "KIYEOK"
@@ -67,39 +59,32 @@ def decompose_jongseong(jongseong: str, /) -> tuple[str, str] | None:
 
     T = TypeVar("T")
 
-    def _create_lookup_table(
-        convert: Callable[[str], T], base: int, end: int
-    ) -> list[T]:
+    def _mklookup(convert: Callable[[str], T], base: int, end: int) -> list[T]:
         return [convert(chr(code)) for code in range(base, end + 1)]
 
-    JAMO_TO_COMPAT_JAMO = _create_lookup_table(
+    CHOSEONG_TO_COMPAT_JAUM = _mklookup(
         jamo_to_compat_jamo,
-        hg.JAMO_BASE,
-        hg.JAMO_END,
+        hg.MODERN_CHOSEONG_BASE,
+        hg.MODERN_CHOSEONG_END,
+    )
+    JONGSEONG_TO_COMPAT_JAUM = _mklookup(
+        jamo_to_compat_jamo,
+        hg.MODERN_JONGSEONG_BASE,
+        hg.MODERN_JONGSEONG_END,
     )
-    COMPAT_JAUM_TO_CHOSEONG = _create_lookup_table(
+
+    COMPAT_JAUM_TO_CHOSEONG = _mklookup(
         compat_jaum_to_choseong,
         hg.MODERN_COMPAT_JAUM_BASE,
         hg.MODERN_COMPAT_JAUM_END,
     )
-    COMPAT_MOUM_TO_JUNGSEONG = _create_lookup_table(
-        compat_moum_to_jungseong,
-        hg.MODERN_COMPAT_MOUM_BASE,
-        hg.MODERN_COMPAT_MOUM_END,
-    )
-    COMPAT_JAUM_TO_JONGSEONG = _create_lookup_table(
+    COMPAT_JAUM_TO_JONGSEONG = _mklookup(
         compat_jaum_to_jongseong,
         hg.MODERN_COMPAT_JAUM_BASE,
         hg.MODERN_COMPAT_JAUM_END,
     )
-    DECOMPOSE_JONGSEONG = _create_lookup_table(
-        decompose_jongseong,
-        hg.MODERN_JONGSEONG_BASE,
-        hg.MODERN_JONGSEONG_END,
-    )
 
-    print(f"JAMO_TO_COMPAT_JAMO = {JAMO_TO_COMPAT_JAMO}\n")
+    print(f"CHOSEONG_TO_COMPAT_JAUM = {CHOSEONG_TO_COMPAT_JAUM}\n")
+    print(f"JONGSEONG_TO_COMPAT_JAUM = {JONGSEONG_TO_COMPAT_JAUM}\n")
     print(f"COMPAT_JAUM_TO_CHOSEONG = {COMPAT_JAUM_TO_CHOSEONG}\n")
-    print(f"COMPAT_MOUM_TO_JUNGSEONG = {COMPAT_MOUM_TO_JUNGSEONG}\n")
     print(f"COMPAT_JAUM_TO_JONGSEONG = {COMPAT_JAUM_TO_JONGSEONG}\n")
-    print(f"DECOMPOSE_JONGSEONG = {DECOMPOSE_JONGSEONG}\n")
diff --git a/ricecake/hangul/_lookup.py b/ricecake/hangul/_lookup.py
@@ -1,4 +1,4 @@
-JAMO_TO_COMPAT_JAMO = [
+CHOSEONG_TO_COMPAT_JAUM = [
     "ㄱ",
     "ㄲ",
     "ㄴ",
@@ -18,155 +18,9 @@
     "ㅌ",
     "ㅍ",
     "ㅎ",
-    None,
-    "ㅥ",
-    "ㅦ",
-    None,
-    None,
-    None,
-    None,
-    "ㅀ",
-    None,
-    "ㅮ",
-    "ㅱ",
-    "ㅲ",
-    None,
-    "ㅳ",
-    "ㅄ",
-    "ㅴ",
-    "ㅵ",
-    None,
-    None,
-    None,
-    "ㅶ",
-    None,
-    "ㅷ",
-    None,
-    "ㅸ",
-    "ㅹ",
-    "ㅺ",
-    "ㅻ",
-    "ㅼ",
-    None,
-    None,
-    "ㅽ",
-    None,
-    None,
-    None,
-    "ㅾ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㅿ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㆀ",
-    None,
-    None,
-    None,
-    None,
-    "ㆁ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㆄ",
-    "ㆅ",
-    "ㆆ",
-    None,
-    "ㅧ",
-    "ㄵ",
-    "ㄶ",
-    None,
-    None,
-    None,
-    "ㅏ",
-    "ㅐ",
-    "ㅑ",
-    "ㅒ",
-    "ㅓ",
-    "ㅔ",
-    "ㅕ",
-    "ㅖ",
-    "ㅗ",
-    "ㅘ",
-    "ㅙ",
-    "ㅚ",
-    "ㅛ",
-    "ㅜ",
-    "ㅝ",
-    "ㅞ",
-    "ㅟ",
-    "ㅠ",
-    "ㅡ",
-    "ㅢ",
-    "ㅣ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㆇ",
-    "ㆈ",
-    None,
-    None,
-    "ㆉ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㆊ",
-    "ㆋ",
-    None,
-    "ㆌ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㆍ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
+]
+
+JONGSEONG_TO_COMPAT_JAUM = [
     "ㄱ",
     "ㄲ",
     "ㄳ",
@@ -194,67 +48,6 @@
     "ㅌ",
     "ㅍ",
     "ㅎ",
-    None,
-    None,
-    None,
-    "ㅦ",
-    "ㅧ",
-    "ㅨ",
-    None,
-    None,
-    None,
-    "ㅩ",
-    None,
-    "ㅪ",
-    None,
-    None,
-    None,
-    None,
-    "ㅫ",
-    None,
-    None,
-    None,
-    "ㅬ",
-    None,
-    "ㅭ",
-    None,
-    None,
-    "ㅮ",
-    "ㅯ",
-    None,
-    "ㅰ",
-    None,
-    None,
-    "ㅱ",
-    None,
-    None,
-    None,
-    "ㅸ",
-    "ㅺ",
-    "ㅼ",
-    None,
-    "ㅽ",
-    "ㅿ",
-    None,
-    None,
-    "ㆀ",
-    None,
-    "ㆁ",
-    "ㆂ",
-    "ㆃ",
-    None,
-    "ㆄ",
-    None,
-    None,
-    None,
-    None,
-    "ㆆ",
-    None,
-    None,
-    None,
-    None,
-    None,
-    "ㅥ",
 ]
 
 COMPAT_JAUM_TO_CHOSEONG = [
@@ -290,30 +83,6 @@
     "ᄒ",
 ]
 
-COMPAT_MOUM_TO_JUNGSEONG = [
-    "ᅡ",
-    "ᅢ",
-    "ᅣ",
-    "ᅤ",
-    "ᅥ",
-    "ᅦ",
-    "ᅧ",
-    "ᅨ",
-    "ᅩ",
-    "ᅪ",
-    "ᅫ",
-    "ᅬ",
-    "ᅭ",
-    "ᅮ",
-    "ᅯ",
-    "ᅰ",
-    "ᅱ",
-    "ᅲ",
-    "ᅳ",
-    "ᅴ",
-    "ᅵ",
-]
-
 COMPAT_JAUM_TO_JONGSEONG = [
     "ᆨ",
     "ᆩ",
@@ -346,33 +115,3 @@
     "ᇁ",
     "ᇂ",
 ]
-
-DECOMPOSE_JONGSEONG = [
-    None,
-    ("ᆨ", "ᆨ"),
-    ("ᆨ", "ᆺ"),
-    None,
-    ("ᆫ", "ᆽ"),
-    ("ᆫ", "ᇂ"),
-    None,
-    None,
-    ("ᆯ", "ᆨ"),
-    ("ᆯ", "ᆷ"),
-    ("ᆯ", "ᆸ"),
-    ("ᆯ", "ᆺ"),
-    ("ᆯ", "ᇀ"),
-    ("ᆯ", "ᇁ"),
-    ("ᆯ", "ᇂ"),
-    None,
-    None,
-    ("ᆸ", "ᆺ"),
-    None,
-    ("ᆺ", "ᆺ"),
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-    None,
-]